Paul D. Isaac

The validation of four ultrametric clustering algorithms

Abstract The present simulation study examined the ability of four hierarchical clustering algorithms to recover the true structure in data sets which satisfied both the ultrametric inequality and the structural model of the clustering procedures. The results indicated that the rank order performance of the four methods differed markedly from the rank order generally found in multivariate normal mixture model studies. The differing rank orders demonstrates a lack of robustness of the algorithms over alternative conceptualizations of cluster structure.

On the determination of appropriate dimensionality in data with error

The study deals with the problem of determining true dimensionality of data-with-error scaled by Kruskals multidimensional scaling technique. Artificial data was constructed for 6, 8, 12, 16, and 30 point configurations of 1, 2, or 3 true dimensions by adding varying amounts of error to the true distances. Results show how stress is affected by error, number of points, and number of dimensions, and indicate that stress and the “elbow” criterion are inadequate for purposes of identifying true dimensionality when there is error in the data. The Wagenaar-Padmos procedure for identifying true dimensionality and error level is discussed. A simplified technique, involving a measure calledConstraint, is suggested.

Heart Rate Variability and Cardiac Response to an Auditory Stimulus

While in a state of quiet sleep, 15 full-term male, 15 full-term female, and 15 premature male infants were presented a single auditory stimulus. Heart rate was continuously monitored and scored for mean and standard deviation in the 10-sec prestimulus period, as well as for peak and trough magnitude of the response and the latencies to these points. The infants tended to show a diphasic response, acceleration followed by rebound deceleration below baseline. Prestimulus mean and variability were inversely related, and significantly predictive of the amount of acceleration. Premature infants showed a high resting heart rate, and a milder response to the stimulus. Also evident were larger accelerations and longer latencies to trough for full-term females.

Parental educational level, gender differences, and seniors' aspirations for advanced study

Many factors affect an undergraduate students postbaccalaureate plans. Academic interests, career-related factors, finances, and various psychological and sociological factors all have been found to be related to interest in advanced education. Few studies, however, are gender-specific in examining the relation of parental educational attainment to their childrens aspirations. Such data would be helpful in the attempts to understand the continued underrepresentation of women in most fields requiring advanced degrees. All seniors at a large Midwestern university with at least a 3.0 cumulative grade point average planning to graduate in spring or summer of 1986 were sent questionnaires dealing with their postgraduation plans. Of these 1900 surveys, 791 (42%) were returned. Results reported here examine the relation of future educational plans to gender of student, educational level of parent, and field of study. A modeling effect not noted in previous literature is discussed.

FACTORS RELATED TO DOCTORAL DISSERTATION TOPIC SELECTION

Although the completed doctoral dissertation represents the culmination of a doctoral program, is often published in journal articles, and serves as the foundation of the early research career of the author, the dissertation has received relatively little attention as a subject of research in its own right. To learn more about how students select their dissertation topics, all doctoral graduates of a major midwestern research university over the one year period from spring quarter, 1986, through winter quarter, 1987, were surveyed. Results indicate that a number of factors influence the selection of a dissertation topic, but not surprisingly the most important is the students own preference. The factors affecting the choice of topics include the preference of the adviser, trends in the field, the likelihood that the research will be published, and the projected benefit of the research on the students job prospects. The importance of the factors vary by field and in some cases by gender. When students select their topic and the extent to which the adviser and dissertation committees are responsive to the student also vary by field.

Dissimilarities as indices of individual perceptual structure

The generality of dissimilarity judgments and of configurations obtained by multidimensional scaling (MDS) as indices of perceptual structure was examined. The Ss made magnitude estimates of dissimilarity of pairs of photos, and MDS configurations were constructed separately for each S. The compatability of dissimilarities (judgments or distances in MDS configurations) with two other tasks Ss performed was then examined. These additional tasks, an “encoding” and a “decoding” task, required the S to first construct a message by making use of a set of four photos and, a week later, to decode the messages. Results obtained were interpreted as indicating a degree of generality of the dissimilarities and supporting a hypothesis of individual differences in perceptual structure for the stimulus set. A “data-purification” hypothesis for MDS was discussed and examined.